In humans, insulin like growth factor 2 (IGF2) mRNA binding proteins (IMPs) have been shown to be poor prognostic indicators in cancer. Work from our lab and others indicate that the two most distantly related members, ZBP1 and IMP2, accomplish this by playing drastically different roles within cells. ZBP1 (IMP1) participates in cellular organization, motility and metastasis and knockout mice are developmentally delayed and embryonic lethal. Interestingly, IMP2 knockout mice display prolonged lifespan and resistance to obesity through upregulation of mitochondrial metabolism. Work from our lab suggests that these cellular effects are mediated by the unique RNAs targets of these highly conserved and highly homologous proteins. This recognition of RNAs by IMP members is dictated by strict rules and highly conserved binding elements within the RNA target sequences. To understand how these proteins utilize their consensus sequences to guide the fate of the cell we propose a number of structural and functional studies. After determining the consensus element for IMP2 I will query the genome to identify targets of IMP2 and compare them to published ZBP1 targets. To determine how the difference in RNA preference is generated between the two proteins, I have used NMR spectroscopy to begin solving the structure of IMP2 bound to its consensus elements. By determining which amino acids of IMP2 interact with each of the binding elements, and comparing to the solved ZBP1 structure, I will understand how these proteins generate target specificity. Directed mutagenesis will then be used to interconvert the binding of each RNA binding protein. To gain mechanistic insight into how IMP2 regulates cellular metabolism, I use my determine target sequences to study its role as a trans-acting factor for mitochondrial RNA localization. A number of studies have isolated mRNAs that are preferentially localized and translated near the surface of the mitochondria (many of which are putative IMP2 targets). Through a combination of super registration and high speed live cell imaging I hope to tease apart the individual contributions of ribosomal translocation and IMP2 towards mRNA localization onto the mitochondrial surface. By understanding if this process is a one step co-translational process or if it is a two step sequential RBP regulated process, we can better understand how translational regulation of mitochondrial proteins can regulate metabolic function, both in healthy and diseased states. I propose a multifaceted approach to understand how the IMP family (and possibly other KH domain containing RBPs) generate sequence specificity through subtle changes in the structure of its RNA recognition element. Our approach will accomplish this by determining targets which IMP2 recognizes and by understanding how these lead to a unique role in metabolic regulation. As IMP family members have been shown to be upregulated in numerous cancers and confer poor prognosis, it is likely that the role of these RBPs in oncogenesis is an important and previously unmet area for investigation.